In [1]:
from __future__ import print_function
#%matplotlib inline
import argparse
import os
import random
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.backends.cudnn as cudnn
import torch.optim as optim
import torch.utils.data
import torchvision.datasets as dset
import torchvision.transforms as transforms
import torchvision.utils as vutils
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.animation as animation
from IPython.display import HTML

# Set random seed for reproducibility
manualSeed = 999
#manualSeed = random.randint(1, 10000) # use if you want new results
print("Random Seed: ", manualSeed)
random.seed(manualSeed)
torch.manual_seed(manualSeed)
Random Seed:  999
Out[1]:
<torch._C.Generator at 0x1a0e52d95b0>
In [2]:
# Root directory for dataset
dataroot = "resized_data"

# Number of workers for dataloader
workers = 2

# Batch size during training
batch_size = 64

# Spatial size of training images. All images will be resized to this
#   size using a transformer.
image_size = 64

# Number of channels in the training images. For color images this is 3
nc = 3

# Size of z latent vector (i.e. size of generator input)
nz = 100

# Size of feature maps in generator
ngf = 64

# Size of feature maps in discriminator
ndf = 64

# Number of training epochs
num_epochs = 100

# Learning rate for optimizers
lr = 0.002

# Beta1 hyperparam for Adam optimizers
beta1 = 0.5

# Number of GPUs available. Use 0 for CPU mode.
ngpu = 1
In [3]:
# We can use an image folder dataset the way we have it setup.
# Create the dataset
dataset = dset.ImageFolder(root=dataroot,
                           transform=transforms.Compose([
                               transforms.Resize(image_size),
                               transforms.CenterCrop(image_size),
                               transforms.ToTensor(),
                               transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)),
                           ]))
# Create the dataloader
dataloader = torch.utils.data.DataLoader(dataset, batch_size=batch_size,
                                         shuffle=True, num_workers=workers)

# Decide which device we want to run on
device = torch.device("cuda:0" if (torch.cuda.is_available() and ngpu > 0) else "cpu")

# Plot some training images
real_batch = next(iter(dataloader))
#real_batch是一个列表
#第一个元素real_batch[0]是[128,3,64,64]的tensor,就是标准的一个batch的4D结构:128张图,3个通道,64长,64宽
#第二个元素real_batch[1]是第一个元素的标签,有128个label值全为0
plt.figure(figsize=(8,8))
plt.axis("off")
plt.title("Training Images")
plt.imshow(np.transpose(vutils.make_grid(real_batch[0].to(device)[:64], padding=2, normalize=True).cpu(),(1,2,0)))

#这个函数能让图片显示
#plt.show() 
Out[3]:
<matplotlib.image.AxesImage at 0x1a0ed1305e0>
In [4]:
# custom weights initialization called on netG and netD
def weights_init(m):
    classname = m.__class__.__name__
    if classname.find('Conv') != -1:
        nn.init.normal_(m.weight.data, 0.0, 0.02)
    elif classname.find('BatchNorm') != -1:
        nn.init.normal_(m.weight.data, 1.0, 0.02)
        nn.init.constant_(m.bias.data, 0)
In [5]:
class Generator(nn.Module):
    def __init__(self, ngpu):
        super(Generator, self).__init__()
        self.ngpu = ngpu
        self.main = nn.Sequential(
            # input is Z, going into a convolution
            nn.ConvTranspose2d( nz, ngf * 8, 4, 1, 0, bias=False),
            #输入100,输出64*8,核函数是4*4
            
            nn.BatchNorm2d(ngf * 8),
            nn.ReLU(True),
            
            # state size. (ngf*8) x 4 x 4
            nn.ConvTranspose2d(ngf * 8, ngf * 4, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ngf * 4),
            nn.ReLU(True),
            
            # state size. (ngf*4) x 8 x 8
            nn.ConvTranspose2d( ngf * 4, ngf * 2, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ngf * 2),
            nn.ReLU(True),
            
            # state size. (ngf*2) x 16 x 16
            nn.ConvTranspose2d( ngf * 2, ngf, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ngf),
            nn.ReLU(True),
            
            # state size. (ngf) x 32 x 32
            nn.ConvTranspose2d( ngf, nc, 4, 2, 1, bias=False),
            nn.Tanh()
            
            # state size. (nc) x 64 x 64
        )

    def forward(self, input):
        return self.main(input)
In [6]:
# Create the generator
netG = Generator(ngpu).to(device)

# Handle multi-gpu if desired
if (device.type == 'cuda') and (ngpu > 1):
    netG = nn.DataParallel(netG, list(range(ngpu)))

# Apply the weights_init function to randomly initialize all weights
#  to mean=0, stdev=0.2.
netG.apply(weights_init)

# Print the model
print(netG)
Generator(
  (main): Sequential(
    (0): ConvTranspose2d(100, 512, kernel_size=(4, 4), stride=(1, 1), bias=False)
    (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (2): ReLU(inplace=True)
    (3): ConvTranspose2d(512, 256, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)
    (4): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (5): ReLU(inplace=True)
    (6): ConvTranspose2d(256, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)
    (7): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (8): ReLU(inplace=True)
    (9): ConvTranspose2d(128, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)
    (10): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (11): ReLU(inplace=True)
    (12): ConvTranspose2d(64, 3, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)
    (13): Tanh()
  )
)
In [7]:
class Discriminator(nn.Module):
    def __init__(self, ngpu):
        super(Discriminator, self).__init__()
        self.ngpu = ngpu
        self.main = nn.Sequential(
            # input is (nc) x 64 x 64
            nn.Conv2d(nc, ndf, 4, 2, 1, bias=False),
            nn.LeakyReLU(0.2, inplace=True),
            # state size. (ndf) x 32 x 32
            nn.Conv2d(ndf, ndf * 2, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ndf * 2),
            nn.LeakyReLU(0.2, inplace=True),
            # state size. (ndf*2) x 16 x 16
            nn.Conv2d(ndf * 2, ndf * 4, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ndf * 4),
            nn.LeakyReLU(0.2, inplace=True),
            # state size. (ndf*4) x 8 x 8
            nn.Conv2d(ndf * 4, ndf * 8, 4, 2, 1, bias=False),
            nn.BatchNorm2d(ndf * 8),
            nn.LeakyReLU(0.2, inplace=True),
            # state size. (ndf*8) x 4 x 4
            nn.Conv2d(ndf * 8, 1, 4, 1, 0, bias=False),
            nn.Sigmoid()
        )

    def forward(self, input):
        return self.main(input)
In [8]:
# Create the Discriminator
netD = Discriminator(ngpu).to(device)

# Handle multi-gpu if desired
if (device.type == 'cuda') and (ngpu > 1):
    netD = nn.DataParallel(netD, list(range(ngpu)))

# Apply the weights_init function to randomly initialize all weights
#  to mean=0, stdev=0.2.
netD.apply(weights_init)

# Print the model
print(netD)
Discriminator(
  (main): Sequential(
    (0): Conv2d(3, 64, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)
    (1): LeakyReLU(negative_slope=0.2, inplace=True)
    (2): Conv2d(64, 128, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)
    (3): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (4): LeakyReLU(negative_slope=0.2, inplace=True)
    (5): Conv2d(128, 256, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)
    (6): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (7): LeakyReLU(negative_slope=0.2, inplace=True)
    (8): Conv2d(256, 512, kernel_size=(4, 4), stride=(2, 2), padding=(1, 1), bias=False)
    (9): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (10): LeakyReLU(negative_slope=0.2, inplace=True)
    (11): Conv2d(512, 1, kernel_size=(4, 4), stride=(1, 1), bias=False)
    (12): Sigmoid()
  )
)
In [9]:
# Initialize BCELoss function
criterion = nn.BCELoss()
4
# Create batch of latent vectors that we will use to visualize
#  the progression of the generator
fixed_noise = torch.randn(64, nz, 1, 1, device=device)

# Establish convention for real and fake labels during training
real_label = 1
fake_label = 0

# Setup Adam optimizers for both G and D
optimizerD = optim.Adam(netD.parameters(), lr=lr, betas=(beta1, 0.999))
optimizerG = optim.Adam(netG.parameters(), lr=lr, betas=(beta1, 0.999))
In [10]:
from __future__ import print_function
import argparse
import os
import random
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.backends.cudnn as cudnn
import torch.optim as optim
import torch.utils.data
import torchvision.datasets as dset
import torchvision.transforms as transforms
import torchvision.utils as vutils
import torch.nn.functional as F
import numpy as np
import matplotlib.pyplot as plt
import torchvision.models as models
import matplotlib.animation as animation
from IPython.display import HTML

from scipy import linalg
from torch.nn.functional import adaptive_avg_pool2d

from PIL import Image

import matplotlib.pyplot as plt
import sys
import numpy as np
import os
# print(os.listdir("../input"))

import time
In [11]:
class InceptionV3(nn.Module):
    """Pretrained InceptionV3 network returning feature maps"""

    # Index of default block of inception to return,
    # corresponds to output of final average pooling
    DEFAULT_BLOCK_INDEX = 3

    # Maps feature dimensionality to their output blocks indices
    BLOCK_INDEX_BY_DIM = {
        64: 0,   # First max pooling features
        192: 1,  # Second max pooling featurs
        768: 2,  # Pre-aux classifier features
        2048: 3  # Final average pooling features
    }

    def __init__(self,
                 output_blocks=[DEFAULT_BLOCK_INDEX],
                 resize_input=True,
                 normalize_input=True,
                 requires_grad=False):
        
        super(InceptionV3, self).__init__()

        self.resize_input = resize_input
        self.normalize_input = normalize_input
        self.output_blocks = sorted(output_blocks)
        self.last_needed_block = max(output_blocks)

        assert self.last_needed_block <= 3, \
            'Last possible output block index is 3'

        self.blocks = nn.ModuleList()

        
        inception = models.inception_v3(weights="IMAGENET1K_V1", progress=True)

        # Block 0: input to maxpool1
        block0 = [
            inception.Conv2d_1a_3x3,
            inception.Conv2d_2a_3x3,
            inception.Conv2d_2b_3x3,
            nn.MaxPool2d(kernel_size=3, stride=2)
        ]
        self.blocks.append(nn.Sequential(*block0))

        # Block 1: maxpool1 to maxpool2
        if self.last_needed_block >= 1:
            block1 = [
                inception.Conv2d_3b_1x1,
                inception.Conv2d_4a_3x3,
                nn.MaxPool2d(kernel_size=3, stride=2)
            ]
            self.blocks.append(nn.Sequential(*block1))

        # Block 2: maxpool2 to aux classifier
        if self.last_needed_block >= 2:
            block2 = [
                inception.Mixed_5b,
                inception.Mixed_5c,
                inception.Mixed_5d,
                inception.Mixed_6a,
                inception.Mixed_6b,
                inception.Mixed_6c,
                inception.Mixed_6d,
                inception.Mixed_6e,
            ]
            self.blocks.append(nn.Sequential(*block2))

        # Block 3: aux classifier to final avgpool
        if self.last_needed_block >= 3:
            block3 = [
                inception.Mixed_7a,
                inception.Mixed_7b,
                inception.Mixed_7c,
                nn.AdaptiveAvgPool2d(output_size=(1, 1))
            ]
            self.blocks.append(nn.Sequential(*block3))

        for param in self.parameters():
            param.requires_grad = requires_grad

    def forward(self, inp):
        """Get Inception feature maps
        Parameters
        ----------
        inp : torch.autograd.Variable
            Input tensor of shape Bx3xHxW. Values are expected to be in
            range (0, 1)
        Returns
        -------
        List of torch.autograd.Variable, corresponding to the selected output
        block, sorted ascending by index
        """
        outp = []
        x = inp

        if self.resize_input:
            x = F.interpolate(x,
                              size=(299, 299),
                              mode='bilinear',
                              align_corners=False)

        if self.normalize_input:
            x = 2 * x - 1  # Scale from range (0, 1) to range (-1, 1)

        for idx, block in enumerate(self.blocks):
            x = block(x)
            if idx in self.output_blocks:
                outp.append(x)

            if idx == self.last_needed_block:
                break

        return outp
    
block_idx = InceptionV3.BLOCK_INDEX_BY_DIM[2048]
model = InceptionV3([block_idx])
model=model.cuda()
In [12]:
def calculate_activation_statistics(images,model,batch_size=128, dims=2048,
                    cuda=False):
    model.eval()
    act=np.empty((len(images), dims))
    
    if cuda:
        batch=images.cuda()
    else:
        batch=images
    pred = model(batch)[0]

        # If model output is not scalar, apply global spatial average pooling.
        # This happens if you choose a dimensionality not equal 2048.
    if pred.size(2) != 1 or pred.size(3) != 1:
        pred = adaptive_avg_pool2d(pred, output_size=(1, 1))

    act= pred.cpu().data.numpy().reshape(pred.size(0), -1)
    
    mu = np.mean(act, axis=0)
    sigma = np.cov(act, rowvar=False)
    return mu, sigma
In [13]:
def calculate_frechet_distance(mu1, sigma1, mu2, sigma2, eps=1e-6):
    """Numpy implementation of the Frechet Distance.
    The Frechet distance between two multivariate Gaussians X_1 ~ N(mu_1, C_1)
    and X_2 ~ N(mu_2, C_2) is
            d^2 = ||mu_1 - mu_2||^2 + Tr(C_1 + C_2 - 2*sqrt(C_1*C_2)).
    """

    mu1 = np.atleast_1d(mu1)
    mu2 = np.atleast_1d(mu2)

    sigma1 = np.atleast_2d(sigma1)
    sigma2 = np.atleast_2d(sigma2)

    assert mu1.shape == mu2.shape, \
        'Training and test mean vectors have different lengths'
    assert sigma1.shape == sigma2.shape, \
        'Training and test covariances have different dimensions'

    diff = mu1 - mu2

    
    covmean, _ = linalg.sqrtm(sigma1.dot(sigma2), disp=False)
    if not np.isfinite(covmean).all():
        msg = ('fid calculation produces singular product; '
               'adding %s to diagonal of cov estimates') % eps
        print(msg)
        offset = np.eye(sigma1.shape[0]) * eps
        covmean = linalg.sqrtm((sigma1 + offset).dot(sigma2 + offset))

    
    if np.iscomplexobj(covmean):
        if not np.allclose(np.diagonal(covmean).imag, 0, atol=1e-3):
            m = np.max(np.abs(covmean.imag))
            raise ValueError('Imaginary component {}'.format(m))
        covmean = covmean.real

    tr_covmean = np.trace(covmean)

    return (diff.dot(diff) + np.trace(sigma1) +
            np.trace(sigma2) - 2 * tr_covmean)
In [14]:
def calculate_fretchet(images_real,images_fake,model):
     mu_1,std_1=calculate_activation_statistics(images_real,model,cuda=True)
     mu_2,std_2=calculate_activation_statistics(images_fake,model,cuda=True)
    
     """get fretched distance"""
     fid_value = calculate_frechet_distance(mu_1, std_1, mu_2, std_2)
     return fid_value
In [15]:
print("Generator Parameters:",sum(p.numel() for p in netG.parameters() if p.requires_grad))
print("Discriminator Parameters:",sum(p.numel() for p in netD.parameters() if p.requires_grad))
Generator Parameters: 3576704
Discriminator Parameters: 2765568
In [16]:
# Training Loop

# Lists to keep track of progress
img_list = []
G_losses = []
D_losses = []
iters = 0

print("Starting Training Loop...")
# For each epoch
for epoch in range(num_epochs):
    # For each batch in the dataloader
    for i, data in enumerate(dataloader, 0):

        ############################
        # (1) Update D network: maximize log(D(x)) + log(1 - D(G(z)))
        ###########################
        ## Train with all-real batch
        netD.zero_grad()
        # Format batch
        real_cpu = data[0].to(device)
        b_size = real_cpu.size(0)
        label = torch.full((b_size,), real_label, device=device).to(torch.float32)
        # Forward pass real batch through D
        output = netD(real_cpu).view(-1)
        # Calculate loss on all-real batch
        errD_real = criterion(output, label)
        # Calculate gradients for D in backward pass
        errD_real.backward()
        D_x = output.mean().item()

        ## Train with all-fake batch
        # Generate batch of latent vectors
        noise = torch.randn(b_size, nz, 1, 1, device=device)
        # Generate fake image batch with G
        fake = netG(noise)
        label.fill_(fake_label)
        # Classify all fake batch with D
        output = netD(fake.detach()).view(-1)
        # Calculate D's loss on the all-fake batch
        errD_fake = criterion(output, label)
        # Calculate the gradients for this batch
        errD_fake.backward()
        D_G_z1 = output.mean().item()
        # Add the gradients from the all-real and all-fake batches
        errD = errD_real + errD_fake
        # Update D
        optimizerD.step()

        ############################
        # (2) Update G network: maximize log(D(G(z)))
        ###########################
        netG.zero_grad()
        label.fill_(real_label)  # fake labels are real for generator cost
        # Since we just updated D, perform another forward pass of all-fake batch through D
        output = netD(fake).view(-1)
        # Calculate G's loss based on this output
        errG = criterion(output, label)
        # Calculate gradients for G
        errG.backward()
        D_G_z2 = output.mean().item()
        # Update G
        optimizerG.step()

        # Check how the generator is doing by saving G's output on fixed_noise
        if (iters % 500 == 0) or ((epoch == num_epochs-1) and (i == len(dataloader)-1)):
            with torch.no_grad():
                fake = netG(noise)
                fake_display = netG(fixed_noise).detach().cpu()
            img_list.append(vutils.make_grid(fake, padding=2, normalize=True))

        iters += 1
    G_losses.append(errG.item())
    D_losses.append(errD.item())     
    fretchet_dist=calculate_fretchet(real_cpu,fake,model) 
    if ((epoch+1)%5==0):
        
        print('[%d/%d]\tLoss_D: %.4f\tLoss_G: %.4f\tFretchet_Distance: %.4f'
                      % (epoch+1, num_epochs,
                         errD.item(), errG.item(),fretchet_dist))
        
        
        plt.figure(figsize=(8,8))
        plt.axis("off")
        pictures=vutils.make_grid(fake_display[torch.randint(len(fake_display), (10,))],nrow=5,padding=2, normalize=True)
        plt.imshow(np.transpose(pictures,(1,2,0)))
        plt.show()
Starting Training Loop...
[5/100]	Loss_D: 1.5857	Loss_G: 1.4020	Fretchet_Distance: 322.2033
[10/100]	Loss_D: 1.5412	Loss_G: 5.5715	Fretchet_Distance: 360.3366
[15/100]	Loss_D: 0.8169	Loss_G: 2.0347	Fretchet_Distance: 366.7939
[20/100]	Loss_D: 0.8832	Loss_G: 4.1900	Fretchet_Distance: 366.7740
[25/100]	Loss_D: 1.3818	Loss_G: 2.2463	Fretchet_Distance: 338.8822
[30/100]	Loss_D: 0.9591	Loss_G: 2.7507	Fretchet_Distance: 280.4865
[35/100]	Loss_D: 1.1999	Loss_G: 1.5716	Fretchet_Distance: 325.0906
[40/100]	Loss_D: 1.0835	Loss_G: 2.3458	Fretchet_Distance: 314.5325
[45/100]	Loss_D: 0.9259	Loss_G: 2.3333	Fretchet_Distance: 269.9121
[50/100]	Loss_D: 0.9678	Loss_G: 2.8504	Fretchet_Distance: 283.0313
[55/100]	Loss_D: 0.7157	Loss_G: 4.0971	Fretchet_Distance: 250.5658
[60/100]	Loss_D: 0.7522	Loss_G: 2.4933	Fretchet_Distance: 256.9414
[65/100]	Loss_D: 0.8513	Loss_G: 2.9043	Fretchet_Distance: 247.3284
[70/100]	Loss_D: 0.5410	Loss_G: 5.7085	Fretchet_Distance: 256.8563
[75/100]	Loss_D: 0.3755	Loss_G: 4.1280	Fretchet_Distance: 266.9063
[80/100]	Loss_D: 0.3625	Loss_G: 4.5934	Fretchet_Distance: 251.0170
[85/100]	Loss_D: 0.2866	Loss_G: 4.9386	Fretchet_Distance: 250.0919
[90/100]	Loss_D: 0.2546	Loss_G: 5.8029	Fretchet_Distance: 253.0431
[95/100]	Loss_D: 0.1142	Loss_G: 5.6917	Fretchet_Distance: 241.5811
[100/100]	Loss_D: 0.2769	Loss_G: 5.3979	Fretchet_Distance: 244.1794
In [17]:
plt.figure(figsize=(10,5))
plt.title("Generator and Discriminator Loss During Training")
plt.plot(G_losses,label="G")
plt.plot(D_losses,label="D")
plt.xlabel("iterations")
plt.ylabel("Loss")
plt.legend()
plt.show()
In [18]:
# Grab a batch of real images from the dataloader
real_batch = next(iter(dataloader))

# Plot the real images
plt.figure(figsize=(15,15))
plt.subplot(1,2,1)
plt.axis("off")
plt.title("Real Images")
plt.imshow(np.transpose(vutils.make_grid(real_batch[0].to(device)[:64], padding=5, normalize=True).cpu(),(1,2,0)))

# Plot the fake images from the last epoch
plt.subplot(1,2,2)
plt.axis("off")
plt.title("Fake Images")
plt.imshow(np.transpose(img_list[-1].cpu(),(1,2,0)))
plt.show()
In [ ]:
 
In [ ]: